Search Results for author:
Found 41 papers, 35 papers with code
Grounded Chain-of-Thought for Multimodal Large Language Models
Despite great progress, existing multimodal large language models (MLLMs) are prone to visual hallucination, greatly impeding their trustworthy applications.
Long-VITA: Scaling Large Multi-modal Models to 1 Million Tokens with Leading Short-Context Accuray
Establishing the long-context capability of large vision-language models is crucial for video understanding, high-resolution image understanding, multi-modal agents and reasoning.
What Kind of Visual Tokens Do We Need? Training-free Visual Token Pruning for Multi-modal Large Language Models from the Perspective of Graph
Recent Multimodal Large Language Models(MLLMs) often use a large number of visual tokens to compensate their visual shortcoming, leading to excessive computation and obvious visual redundancy.
SVFR: A Unified Framework for Generalized Video Face Restoration
In this paper, we propose a novel approach for the Generalized Video Face Restoration (GVFR) task, which integrates video BFR, inpainting, and colorization tasks that we empirically show to benefit each other.
FlashSloth: Lightning Multimodal Large Language Models via Embedded Visual Compression
Despite a big leap forward in capability, multimodal large language models (MLLMs) tend to behave like a sloth in practical use, i. e., slow response and large latency.
Ranked #69 on
Visual Question Answering
on MM-Vet
Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings
In particular, we reveal that visual tokens will stop contributing to reasoning when the text tokens receive enough image information, yielding obvious visual redundancy.
$γ-$MoD: Exploring Mixture-of-Depth Adaptation for Multimodal Large Language Models
In $\gamma$-MoD, a novel metric is proposed to guide the deployment of MoDs in the MLLM, namely rank of attention maps (ARank).
Ranked #170 on
Visual Question Answering
on MM-Vet
Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Models
In this paper, we propose a novel and training-free approach for the effective visual token pruning of MLLMs, termed FitPrune, which can quickly produce a complete pruning recipe for MLLMs according to a pre-defined budget.
Image Captioning via Dynamic Path Customization
This paper explores a novel dynamic network for vision and language tasks, where the inferring structure is customized on the fly for different inputs.
Deep Instruction Tuning for Segment Anything Model
Recently, Segment Anything Model (SAM) has become a research hotspot in the fields of multimedia and computer vision, which exhibits powerful yet versatile capabilities on various (un) conditional image segmentation tasks.
Not All Attention is Needed: Parameter and Computation Efficient Transfer Learning for Multi-modal Large Language Models
In this paper, we propose a novel parameter and computation efficient tuning method for Multi-modal Large Language Models (MLLMs), termed Efficient Attention Skipping (EAS).
Fast Text-to-3D-Aware Face Generation and Manipulation via Direct Cross-modal Mapping and Geometric Regularization
Text-to-3D-aware face (T3D Face) generation and manipulation is an emerging research hot spot in machine learning, which still suffers from low efficiency and poor quality.
Feast Your Eyes: Mixture-of-Resolution Adaptation for Multimodal Large Language Models
Contrary to previous works, we study this problem from the perspective of image resolution, and reveal that a combination of low- and high-resolution visual features can effectively mitigate this shortcoming.
Ranked #156 on
Visual Question Answering
on MM-Vet
Towards Efficient Diffusion-Based Image Editing with Instant Attention Masks
In particular, InstDiffEdit aims to employ the cross-modal attention ability of existing diffusion models to achieve instant mask guidance during the diffusion steps.
Towards Omni-supervised Referring Expression Segmentation
To address this issue, we propose a new learning task for RES called Omni-supervised Referring Expression Segmentation (Omni-RES), which aims to make full use of unlabeled, fully labeled and weakly labeled data, e. g., referring points or grounding boxes, for efficient RES training.
NICE: Improving Panoptic Narrative Detection and Segmentation with Cascading Collaborative Learning
To address this, we introduce two cascading modules based on the barycenter of the mask, which are Coordinate Guided Aggregation (CGA) and Barycenter Driven Localization (BDL), responsible for segmentation and detection, respectively.
Parameter and Computation Efficient Transfer Learning for Vision-Language Pre-trained Models
In this paper, we aim at parameter and computation efficient transfer learning (PCETL) for VLP models.
MMAPS: End-to-End Multi-Grained Multi-Modal Attribute-Aware Product Summarization
Furthermore, we model product attributes based on both text and image modalities so that multi-modal product characteristics can be manifested in the generated summaries.
Systematic Investigation of Sparse Perturbed Sharpness-Aware Minimization Optimizer
Sharpness-Aware Minimization (SAM) is a popular solution that smooths the loss landscape by minimizing the maximized change of training loss when adding a perturbation to the weight.
Approximated Prompt Tuning for Vision-Language Pre-trained Models
Prompt tuning is a parameter-efficient way to deploy large-scale pre-trained models to downstream tasks by adding task-specific tokens.
Adapting Pre-trained Language Models to Vision-Language Tasks via Dynamic Visual Prompting
In addition, we also experiment DVP with the recently popular adapter approach to keep the most parameters of PLMs intact when adapting to VL tasks, helping PLMs achieve a quick shift between single- and multi-modal tasks.
Cheap and Quick: Efficient Vision-Language Instruction Tuning for Large Language Models
To validate MMA, we apply it to a recent LLM called LLaMA and term this formed large vision-language instructed model as LaVIN.
Active Teacher for Semi-Supervised Object Detection
In this paper, we study teacher-student learning from the perspective of data initialization and propose a novel algorithm called Active Teacher(Source code are available at: \url{https://github. com/HunterJ-Lin/ActiveTeacher}) for semi-supervised object detection (SSOD).
Towards End-to-end Semi-supervised Learning for One-stage Object Detection
In addition to this challenge, we also reveal two key issues in one-stage SSOD, which are low-quality pseudo-labeling and multi-task optimization conflict, respectively.
Towards Efficient Visual Adaption via Structural Re-parameterization
Experimental results show the superior performance and efficiency of RepAdapter than the state-of-the-art PETL methods.
Towards Local Visual Modeling for Image Captioning
In this paper, we study the local visual modeling with grid features for image captioning, which is critical for generating accurate and detailed captions.
Towards Real-Time Panoptic Narrative Grounding by an End-to-End Grounding Network
Extensive experiments on the PNG benchmark dataset demonstrate the effectiveness and efficiency of our method.
RefCLIP: A Universal Teacher for Weakly Supervised Referring Expression Comprehension
Based on RefCLIP, we further propose the first model-agnostic weakly supervised training scheme for existing REC models, where RefCLIP acts as a mature teacher to generate pseudo-labels for teaching common REC models.
RefTeacher: A Strong Baseline for Semi-Supervised Referring Expression Comprehension
In this paper, we present the first attempt of semi-supervised learning for REC and propose a strong baseline method called RefTeacher.
Make Sharpness-Aware Minimization Stronger: A Sparsified Perturbation Approach
One of the popular solutions is Sharpness-Aware Minimization (SAM), which smooths the loss landscape via minimizing the maximized change of training loss when adding a perturbation to the weight.
CycleTrans: Learning Neutral yet Discriminative Features for Visible-Infrared Person Re-Identification
Visible-infrared person re-identification (VI-ReID) is a task of matching the same individuals across the visible and infrared modalities.
A Survivor in the Era of Large-Scale Pretraining: An Empirical Study of One-Stage Referring Expression Comprehension
But the most encouraging finding is that with much less training overhead and parameters, SimREC can still achieve better performance than a set of large-scale pre-trained models, e. g., UNITER and VILLA, portraying the special role of REC in existing V&L research.
Towards Lightweight Transformer via Group-wise Transformation for Vision-and-Language Tasks
Despite the exciting performance, Transformer is criticized for its excessive parameters and computation cost.
PixelFolder: An Efficient Progressive Pixel Synthesis Network for Image Generation
Pixel synthesis is a promising research paradigm for image generation, which can well exploit pixel-wise prior knowledge for generation.
SeqTR: A Simple yet Universal Network for Visual Grounding
In this paper, we propose a simple yet universal network termed SeqTR for visual grounding tasks, e. g., phrase localization, referring expression comprehension (REC) and segmentation (RES).
Ranked #11 on
Referring Expression Segmentation
on RefCOCO testB
DIFNet: Boosting Visual Information Flow for Image Captioning
Current Image captioning (IC) methods predict textual words sequentially based on the input visual information from the visual feature extractor and the partially generated sentence information.
Towards Language-guided Visual Recognition via Dynamic Convolutions
Based on the LaConv module, we further build the first fully language-driven convolution network, termed as LaConvNet, which can unify the visual recognition and multi-modal reasoning in one forward structure.
RSTNet: Captioning With Adaptive Attention on Visual and Non-Visual Words
Then, we build a BERTbased language model to extract language context and propose Adaptive-Attention (AA) module on top of a transformer decoder to adaptively measure the contribution of visual and language cues before making decisions for word prediction.
TRAR: Routing the Attention Spans in Transformer for Visual Question Answering
Due to the superior ability of global dependency modeling, Transformer and its variants have become the primary choice of many vision-and-language tasks.
Multi-task Collaborative Network for Joint Referring Expression Comprehension and Segmentation
In addition, we address a key challenge in this multi-task setup, i. e., the prediction conflict, with two innovative designs namely, Consistency Energy Maximization (CEM) and Adaptive Soft Non-Located Suppression (ASNLS).
Generalized Referring Expression Comprehension
Referring Expression
+2
A Real-time Global Inference Network for One-stage Referring Expression Comprehension
Referring Expression Comprehension (REC) is an emerging research spot in computer vision, which refers to detecting the target region in an image given an text description.
